将 3D DataArray 重塑为 2d DataArray

Question

I have xarray dataarray of the following shape.

(10, 666, 999) as ("band", "y", "x") .

I want to get new dataarray into (666*999, 10) shape. How can I solve it?

I tried as ans = x_arr.transpose("y", "x", "band").reshape(666*999, 10)

but it did not work. AttributeError: 'DataArray' object has no attribute 'reshape'

reference, http://xarray.pydata.org/en/stable/generated/xarray.DataArray.html

Answer 1

You can use stack to combine existing dimensions into a new one. In your case it'd be something like darr.stack(xy=["x", "y"])

Here's a reproducible example:

x = xr.tutorial.load_dataset("air_temperature")
print(x.air.shape)
#(2920, 25, 53)

x2 = x.air.stack(point=["lat", "lon"])
print(x2.shape)
#(2920, 1325)